149results about How to "Accurate score" patented technology

The present invention is a system, method, and program product that comprises a computer with a collection of documents to be searched. The documents contain free form (natural language) text. We define a set of labels called QA-Tokens, which function as abstractions of phrases or question-types. We define a pattern file, which consists of a number of pattern records, each of which has a question template, an associated question word pattern, and an associated set of QA-Tokens. We describe a query-analysis process which receives a query as input and matches it to one or more of the question templates, where a priority algorithm determines which match is used if there is more than one. The query-analysis process then replaces the associated question word pattern in the matching query with the associated set of QA-Tokens, and possibly some other words. This results in a processed query having some combination of original query tokens, new tokens from the pattern file, and QA-Tokens, possibly with weights. We describe a pattern-matching process that identifies patterns of text in the document collection and augments the location with corresponding QA-Tokens. We define a text index data structure which is an inverted list of the locations of all of the words in the document collection, together with the locations of all of the augmented QA-Tokens. A search process then matches the processed query against a window of a user-selected number of sentences that is slid across the document texts. A hit-list of top-scoring windows is returned to the user.

A karaoke scoring apparatus includes a memory element, a feature extraction element, a similarity measurement element, and a scoring element. The karaoke system includes a reference audio input to be compared with a target audio input for giving a score. The reference audio input and the target audio input are sampled respectively and are transformed sequentially to plural frames of reference sampling signals and plural frames of target sampling signals. The memory element is used for storing the frame of reference sampling signal and the frame of target sampling signal. The feature extraction element is used for performing autocorrelation calculation on the frame of reference sampling signal and the frame of target sampling signal. The similarity measurement element is used for generating a similarity result. The scoring element is used for calculating the similarity results corresponding to the plural frames of sampling signals to output final score.

A method for diagnosing, assessing, and determining of the efficacy of a treatment regimen for chronic rhinosinusitis is presented. The method comprises collecting subjective information from a patient or patient population in the form of patient entries into a daily diary, wherein the daily diary comprises the patient responses to questions directed to the presence or absence of a group of clinical signs and symptoms related to chronic rhinosinusitis. Based upon the patient responses in the daily diaries, certain of the clinical signs and symptoms are identified for use in the generation of a scoring tool useful in the determination and assessment of the efficacy of a treatment regimen for chronic rhinosinusitis, as well as in the clinical determination of the severity of the symptoms related to chronic rhinosinusitis.

The invention discloses a multidimensional information fusion-based comprehensive e-commerce product scoring method. The method comprises the following steps of: obtaining multidimensional informationsuch as shop information, sales volume information and comment text information of e-commerce products; data preprocessing: carrying out data cleaning and data conversion on numerical type data, andcarrying out word segmentation and part-of-speech tagging on comment texts; mining the multidimensional information: carrying out data reduction and principal component regression analysis on the shopinformation and the commodity sales volume information to obtain shop information indexes and commodity sales volume indexes, carrying out emotion analysis on the comment texts, and obtaining a product feature score radar map through a quantification method and a clustering method; and commodity total score calculation: designing a fusion function and calculating a commodity total score. The method can be applied to commodity information-based commodity recommendation systems, is capable of efficiently and conveniently recognizing high-quality commodities so as to ensure that the designed recommendation systems are more rapid and correct.

The invention discloses an automatic cloth defect point detection and recognition device and method based on machine vision. The method includes the steps of taking the picture of a piece of defect-point-free cloth through a camera to serve as a standard image, saving the standard image into a computer, extracting the characteristic parameters of the standard image through an image processing method based on the combination of wavelet conversion and a protection method, taking the picture of a piece of to-be-detected cloth in real time through the camera, extracting the characteristic parameters, setting the accuracy range according to the requirements, comparing the to-be-detected image with the standard image and analyzing the to-be-detected image and the standard image, judging whether defect points exist or not according to whether the characteristic parameters of the to-be-detected image are within the accuracy range or not, saving the image with the defect points if the defect points exist, recognizing the defect points in the image containing the defect points through an image processing method based on optimized threshold value segmentation, completing classifying and grading of the defect points, and conducting level evaluation on the whole cloth. The device and the method overcome the defects that traditional manual visual measurement is low in speed and accuracy and high in detection omission rate and false drop rate, the labor intensity of manual defect point classification is lowered, and the device and the method have the advantage of being accurate in grading.

The invention provides an anti-fraud data analysis method and system. The method comprises the steps of obtaining a data mart and a graph database from a back-end database, and sending the data mart and the graph database to an analysis model, wherein the data mart and the graph database are generated according to pre-collected basic data on fraud; analyzing the data mart and the graph database bythe analysis model to obtain an analysis result; and outputting the analysis result to a front end and performing display. An individual fraud risk identification model is combined with a group fraudrisk identification model; and an anti-fraud system is completer. Supervised learning, semi-supervised learning and unsupervised learning are integrated; the data is processed; the identification accuracy is improved; and the model is higher in updating speed and higher in adaptability. Meanwhile, an anti-fraud real-time scoring system is established; the scoring calculation is more accurate; andan anti-fraud analysis result is more intuitive to observe. A visual user relationship map also can intuitively display the analysis result, so that the user experience is good.

The invention discloses a fixture used for milling and boring a support. The fixture comprises a station I fixture body, a station II fixture body and a bottom plate, wherein the station I fixture body comprises a first positioning mechanism, a first clamping mechanism and a center positioning mechanism for strutting a workpiece outwards; the center positioning mechanism comprises a supporting seat, two pressing plates I, an angle pole, a guide screw and a driving device, wherein the two pressing plates I are symmetrically arranged at both ends of the supporting seat; the driving device comprises a gear mechanism and a driving oil cylinder; the gear mechanism comprises a double-row rack, gears II and a single-row rack, wherein the gears II are respectively meshed with the double-row rack, and the single-row rack is meshed with the gear II; one end of the angle pole is connected with the pressing plates I; the other end of the angle pole is connected with the single-row rack. According to the fixture disclosed by the invention, clamping can be performed rapidly, and is steady; the work support is not easy to deform, and therefore, the machining precision is improved; moreover, when the support of a symmetrical structure is milled, the center of the support can be rapidly and accurately positioned; moreover, the fixture can be suitable for volume production.

The invention discloses a method for automatically establishing back-of-book indexes of a book based on book contents. The method comprises the following steps: first, analyzing text in a digital book; taking a chapter as a unit, and performing part-of-speech tagging on the text by using a natural language processing tool to obtain a part-of-speech array; matching by utilizing a high-frequency part-of-speech rule, and extracting candidate phrases; then, classifying to obtain the phrases to serve as candidate index terms by using a support vector machine algorithm by utilizing semantic and grammatical characteristics; calculating the similarity between the candidate index terms and the field corresponding to the book to serve as termhood; calculating information amounts, term frequency, point mutual information and encyclopedia key values to obtain an index degree; calculating a title offset distance, a candidate index term proportion and interestingness to obtain a context weight value; finally, combining the termhood, the indexing degree and the context weight value to obtain an index score, and obtaining the book index terms through limited sequencing. According to the method, the indexes can be added to the book which does not have the back-of-book terms, and the readability and the searchability of the book are improved.

An embodiment of the invention discloses a method and device to determine a rating matrix based on a convolutional neural network. The method comprises: using a pre-trained convolutional neural network to process acquired document information, and determining a target characteristic vector; establishing an initial rating matrix according users' rating information for a commodity, wherein blank items where users do not make rating may occur in the initial rating matrix; processing the initial rating matrix according to the target characteristic vector and a matrix-decomposition-based latent semantic model, processing the initial rating matrix, and filling the blank items in the initial rating matrix to obtain a target matrix. By using the convolutional neural network, it is possible to recognize context of document information of a commodity, text information of the commodity from users can be better mined, and the influence due to data sparsity problem can be weakened. By substitutingthe target characteristic vector into the matrix-decomposition-based latent semantic model, predicted ratings in the target matrix can be more accurate.

The invention discloses an online multi-target tracking method and system based on a twin network and long and short-term clues, and belongs to the field of multi-target tracking. The system comprises: a twin network module configured to perform cross-correlation of a tracking target template and a search area to obtain a response graph and obtain a preliminarily predicted tracking trajectory of each target; a correction module which is used for combining the preliminary track and the observation frame together and correcting the pedestrian frame through a pedestrian regression network; a dataassociation module which is used for calculating the similarity between the tracking tracks and the observation pedestrians, respectively extracting long-term and short-term clues of the tracking tracks and the observation pedestrians and fusing the long-term and short-term clues so as to calculate the similarity, and allocating a corresponding observation pedestrian box to each tracking track; and a track post-processing module which is used for updating, supplementing and deleting the tracking track to complete the tracking of the current frame. According to the method, the problems of apparent feature fusion, pedestrian interaction occlusion and large-scale change in a multi-target tracking task are perfected, the accuracy is improved, and the problem of feature misalignment is relieved.

The invention is applicable to computer aided instruction for engineering drawing. According to different users, the method decomposes relevant functions into different module units including management module based on network, kernel module for judging credit based on detailed analyzing knowledge points, module for students to submit homework, module for completing homework and querying feed back, and networked module for managing titles of engineering drawing. Under specific requirement, the system can implement specific rules for judging credits, possessing better flexibility. According to presetting rules, the method can mark highlighted tags on error positions in graph. Through colors, linewidth, line type, and text discrition, taking full advantage of graphical display, the invention realizes adding artificial comment automatically.

The invention provides a karaoke scoring method based on voice matching. In a singing process, an audio frequency of an original singer is extracted. Simultaneously sound of a user is continuously extracted, and an audio characteristic of the user is calculated. Furthermore the audio characteristic of the user is compared with the audio characteristic of the original singer sentence by sentence. Scoring is performed according to different weights. According to the karaoke scoring method, factors of tone, tone quality, volume, rhythm, time length, etc. are not considered; and sound essence is considered. The essential characteristic of the sound is considered. The spectral characteristic of the sound is analyzed. The factors can be comprehensively considered, and a corresponding calculating is performed, thereby obtaining a more comprehensive and more accurate scoring result.

The invention provides a speech recognition method, a device and a storage medium, which relate to the computer technical field. The method comprises the following steps: acquiring at least two candidate recognition results and scores thereof of speech information to be recognized provided by a current user; extracting key words of preset type from each candidate recognition result according to preset keyword extraction rules; using the extracted key words and the scores of the candidate recognition results, the scores of each candidate recognition result are corrected, and the final recognition result of the speech information to be recognized is determined according to the correction result. As that embodiment of the present application revise the score of the candidate recognition result by using the key word extracted from the candidate recognition result, and the key word is used as a correcting factor, each candidate recognition result can obtain more accurate score, and more accurate recognition results can be screened according to the corrected score.

The invention discloses a movie marking prediction method based on implicit bias and interest of friends. The method includes the steps of constructing a marking prediction model integrating personal bias with the interest of the friends according to movie marking logs of users and friend list data of the users, conducting model training to obtain bias matrixes of the users on movies, user implicit feature matrixes and article implicit feature matrixes, generating a friend interest implicit feature matrix through combination with the friend list data and the user implicit feature matrixes, generating a user-article prediction marking matrix according to the user implicit feature matrixes, the article feature matrixes, user-article bias matrixes and the friend interest implicit feature matrix, and finally generating a recommendation list for the users. Through the movie marking prediction method based on the implicit bias and the interest of the friends, marking bias of the users and article preferences of the friends are overall considered, and the recommendation quality is improved.

The embodiment of the invention discloses a named entity recognition method, which comprises the following steps: when an entity in a to-be-recognized text needs to be recognized, obtaining a word vector of a word segmentation entry in the to-be-recognized text; and according to the word vector of the word segmentation entry and the entity recognition model, determining a first score of each typeof label corresponding to the word segmentation entry; Respectively calculating a first matching score between the feature vector of the word segmentation entry and the label vector of each type of label, wherein the first matching score reflects the possibility that the word segmentation entry has each type of label; according to the first score and the first matching score, obtaining a second score of each type of label corresponding to the word segmentation entry, and adding the score of the label of the word segmentation entry in the entity dictionary on the basis of the first score. By combining the first matching score on the output layer of the entity recognition model and adding the score of the label of the word segmentation entry on the basis of the first score, the influence ofthe entity dictionary on calculation of the score of each type of label is enhanced, so that the calculated score is more accurate, and more entities can be recalled.

The invention provides an information recommendation method and device, electronic equipment and a storage medium, and the method comprises the steps: receiving a user request sent by a user terminal,the user request carrying user information; acquiring multiple pieces of recommendation information corresponding to the user request according to the user information; according to the service attribute corresponding to each piece of recommendation information, sorting the plurality of pieces of recommendation information to obtain a recommendation list, wherein the service attribute representsthe service recommendation degree of the corresponding recommendation information; and sending the recommendation list to the user terminal. According to the embodiment of the invention, after a userrequest is received, corresponding recommendation information is obtained according to the user information, service attributes related to the recommendation information are used; multiple pieces of recommendation information are reordered to obtain the recommendation list, and the recommendation list is returned to the user terminal, so that the user can obtain interested information according tothe recommendation list, and meanwhile, the user can pay more attention to the information pushed by the front business in the recommendation list, and development of business promotion is facilitated.

The invention discloses a recommendation method based on collaborative filtering. The recommendation method comprises: calculating the similarity between any two projects; constructing a project similarity matrix according to the similarity between the any two projects; obtaining a nearest neighbor set of each project according to the project similarity matrix; calculating a predicated score of each project predicated by a user according to the nearest neighbor set; calculating the similarity of any two users according to the predicated score of each project and the confidence degree of the predicated score; calculating a nearest neighbor set of each user according to the similarity of the any two users; according to the nearest neighbor set of each user, scoring projects, which are not scored, of each user, and generating a recommendation set of the user according to a scoring result. By utilizing the reliability of nearest neighbor collaborative filtering recommendation, the reliability of complementary data is improved on the basis of the complementary data when data sparsity is not enough, so that the collaborative filtering can play a better role.

The invention relates to the field of bioengineering and tumor markers, in particular to a marker for locally advanced esophageal squamous cell carcinoma prognosis and an application of the marker. The marker is a combined marker consisting of one or more of miR-135b-5p, miR-139-5p, miR-29c-5p and miR-338-3p, and locally advanced esophageal squamous cell carcinoma prognosis is predicted by detecting expression levels of the four miRNA in tumor tissue and performing calculation according to a formula (0.4690*miR-135b-5p expression level)+(0.3839*miR-139-5p expression level)+(0.1733*miR-29c-5p expression level)+(0.3368*miR-338-3p expression level). The combined marker has the advantages of good stability and high sensitivity and specificity and can more accurately and more valuably evaluateprognosis of patients than traditional clinical pathological factors such as TNM (tumor-node-metastasis) staging and the like. On one hand, the molecular marker related to esophageal squamous cell carcinoma prognosis is provided, on the other hand, an esophageal squamous cell carcinoma prognosis prediction model is established, so that individual treatment of esophageal squamous cell carcinoma isrealized, the comprehensive treatment level of the esophageal squamous cell carcinoma is improved, the life quality of esophageal squamous cell carcinoma patients is improved, and the lifetime of theesophageal squamous cell carcinoma patients is prolonged.

One or more embodiments of the invention disclose a score generating method and device, a storage medium and electronic equipment. The method comprises the steps of collecting an input evaluation voice set; obtaining an evaluation text corresponding to the evaluation voice set; comparing the evaluation text with a sample text; obtaining difference text information between the evaluation text and the sample text; based on the difference text information, scoring the evaluation voice set; generating a score corresponding to the evaluation voice set. By means of the one or more embodiments, the voice scoring accuracy can be improved.

The invention relates to a machine vision technology detection method based on depth learning, which relates to the technical field of industrial detection. A large number of defective product pictures are collected as samples by an industrial camera under a good industrial light source; These sample pictures are calibrated by software to mark the position of the defects concerned and the types ofdefects. The sample pictures marked with the defect types are used as the material hook part depth learning network model, and the depth learning network model is trained; The trained in-depth learning network model is imported into machine vision system, so as to identify various defects on the spot, and complete the sorting of defective products with industrial automation equipment. To avoid missed detection and false detection, new samples can be learned to improve the accuracy, adaptability and stability, and realize the expert system to meet the needs of various complex scenarios of industrial detection.

The invention is suitable for the field of computer evaluation, and provides a voice evaluation method and device. The method comprises the steps: obtaining a to-be-evaluated audio; extracting a characteristic quantity of the audio; generating a score corresponding to the characteristic quantity according to a preset scoring standard and the characteristic quantity of the audio; generating and displaying a radar map corresponding to the score according to the score of the characteristic quantity, so as to enable an evaluator to score the to-be-evaluated voice according to the radar map. According to the invention, the method and device employs a mode of man-machine combination, enables the evaluator to be able to accurately give a score through generating the visual radar map, and reduces the scoring fault of the voice evaluation.

The invention discloses an automatic extraction and intelligent scoring system for handwritten Chinese characters or components and strokes. The automatic extraction and intelligent scoring system comprises an automatic extraction module for calligraphy practicing book images, an automatic extraction module for single Chinese characters, a Chinese character recognition module and a Chinese character scoring module. The automatic extraction module for calligraphy practicing books carries out preprocessing, image capture, correction, error reporting and format judgment on uploaded handwritten Chinese character photos, and the accuracy and the automation effect of image extraction are improved through the complete process. The automatic extraction module for single Chinese characters comprises a special processing method for pencil character extraction, so that a pencil character image can be effectively extracted; the established deep learning recognition model can quickly recognize handwritten Chinese characters, including space recognition; and Chinese character scoring adopts a structure+content comprehensive scoring method, the structure is the length and width values of Chinesecharacters, and the content is determined through cosine similarity. According to the method, the handwritten character pictures which are randomly uploaded can be automatically extracted and scored,and the method is suitable for low-age students who begin to learn Chinese characters and is greatly helpful for daily calligraphy practicing of calligraphy enthusiasts.

The invention relates to an audio signal scoring method, device, electronic equipment and a computer storage medium, which belong to the field of audio scoring. First original singing human voice signals separated from original singing signals are subjected to noise suppression processing, second original singing human voice signals with less residual accompaniment can be acquired, the second original singing human voice signals with less residual accompaniment are thus more accurate, the differences between the second original singing human voice signals and target audio signals can be reduced, and the score of the target audio signals is further more accurate.

The embodiment of the invention discloses a method and a device for acquiring a target media file and a storage medium. The method comprises the following steps: loading media data of a plurality of media files; based on the media data, feature extraction is conducted on all the media files, multiple feature values of all the media files are obtained, and the feature values comprise the feature values of the attribute features of the media files and the feature values of the statistical features of the media files; obtaining the score of each media file based on the plurality of characteristicvalues of each media file and the mapping relation between the characteristic values and the scores of the media files; and selecting a preset number of media files in the plurality of media files asa target media file based on the score of each media file.

The invention discloses a medical institution evaluation method and device and a computer readable storage medium which are applied to the field of intelligent recommendation. The method comprises thesteps of obtaining the attribute of a to-be-evaluated target medical institution; determining a plurality of evaluation indexes corresponding to the attributes according to the attributes of the target medical institution; acquiring research data used for indicating importance degrees of the plurality of evaluation indexes; determining weights respectively corresponding to the plurality of evaluation indexes according to the investigation data; evaluating the score of the target medical institution under each evaluation index to obtain the scores of the target medical institution under the plurality of evaluation indexes; and calculating the comprehensive score of the target medical institution by using the weights corresponding to the plurality of evaluation indexes and the scores of thetarget medical institution under the plurality of evaluation indexes. According to the application, the weights corresponding to the evaluation indexes are reasonably set according to the investigation data, and therefore the accurate and reasonable evaluation is conducted on the target medical institution.

The invention provides a teacher teaching quality evaluation system and method, and belongs to the technical field of teaching. The problem that an existing evaluation result is not accurate is solved. The teacher teaching quality evaluation system comprises a database, a web server and a fingerprint attendance server used for obtaining attendance information of students and teachers, the web server and the fingerprint attendance server are connected with the database, and the web server comprises a teacher scoring module, a teacher evaluation module and a teacher evaluation module; the fingerprint attendance management module is used for managing attendance information, stored in the database, of students and teachers; the score calculation module is used for calculating to obtain a student evaluation average score, a supervision evaluation average score, a management evaluation average score and a teacher mutual evaluation average score; and the evaluation result generation module isused for calculating and generating a final score of each course teacher according to the preset weight coefficients of different proportions in combination with the attendance information of the teachers. According to the invention, accuracy of teaching quality evaluation can be realized.

The present invention provides methods for individualizing chemotherapy, and particularly methods for individualizing neoadjuvant chemotherapy. The present invention provides methods for predicting a cancer patient's response to neoadjuvant chemotherapy, including assessing the probability of a positive response upon treatment with candidate agents prior to surgery. In various aspects, the invention involves culturing a monolayer of malignant cells from an explant of a patient's biopsy specimen, such as a transcutaneous biopsy-sized specimen, and testing the malignant cells for resistance or sensitivity to one or a plurality of candidate agents for neoadjuvant therapy. In other aspects, the invention provides methods for accurately scoring and interpreting such assays, and discloses in vitro chemoresponse results that are predictive of a patient's pathological complete response (pCR) upon receiving the corresponding treatment regimen.

The invention discloses a singing scoring method based on lyric and voice alignment. The singing scoring method comprises the following steps in sequence: song recording; voice accompaniment separation and noise removal; extraction of fundamental tone frequency and amplitude; alignment of lyrics with voice by taking sentences as units; segmentation of the fundamental tone frequency of each character in the aligned voice; calculation of a fundamental tone frequency similarity score; calculation of a rhythm score according to the duration of each sentence of user voice and standard voice and thestarting and ending time of each character; normalization of the amplitudes of the user voice and the standard voice; calculation of an amplitude similarity score; multiplying of the fundamental tonefrequency score, the rhythm score and the amplitude score by a weight coefficient and adding, and calculation of a comprehensive score of a song. According to the singing scoring method in the invention, the influence of accompaniment and noise on voice evaluation is reduced; label information of the lyrics is reasonably utilized, so that evaluation of the fundamental tone frequency and rhythm ofthe user is more accurate; and user songs are evaluated in multiple aspects, so that song scoring results are more objective and comprehensive.

The invention discloses an armed mutual fencing entertainment sport high frequency wireless scoring system and method; the high frequency wireless scoring system comprises at least two wireless identification controllers respectively worn on corresponding armed sportsmen entering the field, a USB wireless host arranged in the field circumference, a computer host arranged on the field circumference, and an LED display arranged on the field circumference. A conventional wired scoring mode may restrain the free moves of the armed mutual fencing people, the novel system and method can solve said problems, can automatically identify equipment ID and different valid hit parts, can respectively automatically accumulate corresponding scores of the equipment hit parts, so the armed mutual fencing entertainment sport can be more flexibly carried out, and the results can be effectively recorded and displayed; the novel system and method can score multi-player mutual fencing entertainment sport.